This article is more than 1 year old
Qantas gets rocked by Amadeus
Time leaping blamed for airline mayhem
Airlines Qantas and Virgin Australia went into temporary meltdown on Sunday morning (EST) when the Amadeus ALTEA platform which they both use was affected by unspecified technical issues.
The global glitch left passengers unable to check in across domestic and international flights with staff having to resort to manual processes and cancellations and delays ensuing for several hours.
Qantas took to Twitter to state “global issues with ALTEA system affecting check-in. Issue resolved and ports will be back online shortly. We're sorry for the inconvenience.”
It is the second time in six months that Qantas has had to grapple with Amadeus issues. Back in January the ALTEA system crashed for around three hours affecting its 100 string network of airlines including British Airways, Qantas, and Cathay Pacific.
At the time Amadeus said the massive fail was due to a “network change” and upgrade at its data centre enterprise network in Germany.
Last November, another Amadeus outage had Qantas staff issuing hand-written boarding cards.
In the latest incident, Amadeus confirmed that there were problems with the ALTEA system but would not define them. They were also committed to “taking any appropriate steps to avoid this reoccurring”.
The issue is being widely attributed to the "leap second", the one-second time adjustment for atomic clocks so they can synchronize with clocks based on the Earth's rotation.
Web sites such as FourSquare, and LinkedIn were reportedly affected while more seasoned players like Google had developed a ‘leap smear’ solution to circumvent the issue.
Science commentator Karl Kruszelnicki told ABC radio this morning the leap second problem triggered a security feature that meant Amadeus rejected Qantas transactions. Transactions from Qantas arrived with the wrong time-stamp, which was interpreted by Amadeus as indicating that the transaction may have been intercepted by a man-in-the-middle attack.
The Register has not been able to confirm this explanation for the outage. ®